From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Date: Mon, 29 Jun 2015 06:42:46 +0200 From: Linus =?utf-8?Q?L=C3=BCssing?= Message-ID: <20150629044246.GD2527@odroid> References: <1434893777-17618-1-git-send-email-linus.luessing@c0d3.blue> <2472003.5ybaWVxMcq@voltaire> <20150628035913.GB2398@odroid> <5065799.yHsM5BUa9O@voltaire> <20150629025248.GC2527@odroid> MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: <20150629025248.GC2527@odroid> Subject: Re: [B.A.T.M.A.N.] [PATCH 1/1] batman-adv: Always flood IGMP/MLD reports Reply-To: The list for a Better Approach To Mobile Ad-hoc Networking List-Id: The list for a Better Approach To Mobile Ad-hoc Networking List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Marek Lindner Cc: b.a.t.m.a.n@lists.open-mesh.org On Mon, Jun 29, 2015 at 04:52:48AM +0200, Linus Lüssing wrote: > The alternative to bumping the version number would be to introduce > another flag into the mcast tvlv and keep counters and lists similar > to the num_disabled or want_* ones. So maybe about 80 more lines of > code and additional state and complexity to maintain. Another alternative I just had to think about would be to register both a v1 and v2 TVLV container while only registering a handler for v2. That'd add just two or four extra lines of code with the same result. The disadvantage would be a few bytes extra overhead for dragging a v1 TVLV in the OGMs along. Actually, I think I'd kinda like that as it has a reasonable benefit-to-cost ratio. What do others think? (And maybe such legacy multicast v1 TVLV could then safely be obsoleted/removed in four years or after evaluating whether someone still needs it.)